[gmx-users] defining new residues?

[Mark Abraham]

scott.t.milner at exxonmobil.com wrote:
> Hello -- I am interested in using gromacs to simulate polymers.  Apparently
> the .rtp files within gromacs are defined for protein residues, not typical
> monomers being used in polymers.  I don't know anything about proteins, but
> I would think that there is a strong analogy between proteins and polymers,
> in that linear polymers are composed of a succession of monomers (often,
> just the same monomer repeated), joined by single covalent bonds.  Typical
> monomers I would be interested in are methylene (-CH2-), propylene
> (-CH(CH3)CH2-), styrene (-CH(C6H5)CH2-), and so forth.  It would be very
> convenient if I could add these structures as new "residues" to the
> definitions of some of the force fields (e.g., OPLSAA and GMX), so that I

This would be possible, however it would be unwise to extrapolate that a 
force field for peptide simulations is suitable for lipid-polymer 
simulations without serious evidence, such as previous studies, or the 
force field having been developed with this use in mind. Look in the 
polymer literature and see what force fields are used there. Simplest 
will be to use the software and the force fields described there, rather 
than choose GROMACS and have to work something in to fit. There are 
polymer force fields that work with GROMACS, I just don't know what they 
are...

> could use pdb2gmx to generate topology files for a simulation, without
> having to generate the topology files by hand.  Is there some documentation
> that describes the structure of the various files associated with a force
> field, and what information would then be required for defining new
> residues?

Yep. Chapters 4 and 5 of the manual. They're pretty good, but not 
perfect. I'd recommend playing around doing some simulations to see how 
the pieces fit together before tinkering seriously with force fields.

Mark